Laurae (version 0.0.0.9001)

loss_MAE_xgb: Mean Absolute Error (xgboost function)

Description

This function computes for xgboost's obj function the Mean Absolute Error loss (MAE) gradient and hessian per value provided preds and dtrain.

Usage

loss_MAE_xgb(preds, dtrain)

Arguments

preds

The predictions.

dtrain

The xgboost model.

Value

The gradient and the hessian of the Absolute Error per value in a list.

Details

Supposing: \(x = preds - labels\) Loss Formula : \(abs(x)\) Gradient Formula : \(sign(x)\) Hessian Formula : \(0\)